Uneven terrain exacerbates the deficits of a passive prosthesis in the regulation of whole body angular momentum in individuals with a unilateral transtibial amputation

Abstract

Background: Uneven ground is a frequently encountered, yet little-studied challenge for individuals with amputation. The absence of control at the prosthetic ankle to facilitate correction for surface inconsistencies, and diminished sensory input from the extremity, add unpredictability to an already complex control problem, and leave limited means to produce appropriate corrective responses in a timely manner. Whole body angular momentum, L, and its variability across several strides may provide insight into the extent to which an individual can regulate their movement in such a context. The aim of this study was to explore L in individuals with a transtibial amputation, when challenged by an uneven surface. We hypothesized that, similar to previous studies, sagittal plane L would be asymmetrical on uneven terrain, and further, that uneven terrain would evoke a greater variability in L from stride to stride in individuals with amputation in comparison to unimpaired individuals, due to a limited ability to discern and correct for changing contours beneath the prosthetic foot.

Methods: We examined sagittal plane L in ten individuals with a unilateral transtibial amputation and age- and gender- matched control participants walking on flat (FT) and uneven (UT) treadmills. The average range of L in the first 50% of the gait cycle (L_R), the average L at foot contact (L_C) and their standard deviations (vL_R, vL_C) were computed over 60 strides on each treadmill.

Results: On both surfaces we observed a higher L_R on the prosthetic side and a reduced L_C on the sound side (p < 0.001) in the amputee cohort, consistent with previous findings. UT invoked an increase in L_C (p = 0.006), but not L_R (p = 0.491). vL_R, and vL_C were higher in individuals with amputation (p < 0.001, p = 0.002), and increased in both groups on UT (p < 0.001).

Conclusions: These findings support previous assertions that individuals with amputation regulate L less effectively, and suggest that the deficits of the prosthesis are exacerbated on uneven terrain, potentially to the detriment of balance. Further, the results indicate that a greater demand may be placed on the unaffected side to control movement.

Keywords: Amputees; Angular momentum; Biomechanics; Gait; Motor learning; Passive prostheses; Uneven terrain.

Fig. 1

Sagittal plane rotational dynamics. (a) Whole body angular momentum, L, as the summed momenta of the individual segments, i, illustrated in the sagittal plane (b) Influence of uneven ground on passive prosthesis motion and segmental rotations

Fig. 2

Uneven terrain treadmill. Manually shaped wooden slats were affixed to the belt of a standard treadmill to provide the uneven surface

Fig. 3

Whole body sagittal plane angular momentum, L. Values normalized to mass, height and speed, and time normalized to the right stride (dimensionless). Values are positive in the clockwise direction. Individual participant profiles: (a) prosthetic side gait cycle of 65 yr. old male with right unilateral amputation (b) matched prosthetic side gait cycle of 64 yr. old male with no amputation. Mean of 60 strides ±1 standard deviation. FT – flat terrain, UT – Uneven terrain. L_C – Value of L at foot contact; L_R – Range of L over first 50% of the gait cycle (within non-shaded region)

Fig. 4

Average (mean) sagittal plane angular momentum. Values (a) at foot contact, (L_C) and (b) during ipsilateral stance phase (L_R), normalized to mass, height and speed (unitless). Individuals with (n = 10) and without (n = 10) amputation walking on flat (FT; blue bars) and uneven (UT; gray bars) terrain. Solid and dashed bars represent prosthetic and sound (or matched prosthetic and sound) limbs respectively. Pairwise comparisons: ‘*’ significant difference between limbs (prosthesis vs sound); ‘¥’ significant difference between groups (amputation vs no impairment); all at p = 0.05

Fig. 5

Variability (standard deviation) of sagittal plane angular momentum. Values (a) at foot contact (vL_C) and (b) during ipsilateral stance phase (vL_R), normalized to mass, height and speed (dimensionless). Individuals with (n = 10) and without (n = 10) amputation walking on flat (FT; blue bars) and uneven (UT; gray bars) terrain. Solid and dashed bars indicate prosthetic and sound (or matched prosthetic and sound) limbs respectively. Pairwise comparisons: ‘*’ significant difference between limbs (prosthesis vs sound); ‘¥’ significant difference between groups (amputation vs no impairment); ‘₼’ significant difference between terrain conditions (flat versus uneven); all at p = 0.05